New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

#

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

Jump to bottom

The learned transformation layers are not used when testing? #14

Open

chmxu opened this issue May 10, 2020 · 1 comment

chmxu commented May 10, 2020

Hi.

It seems that the scale and shift are omitted in all feature transformation layers when test.
https://github.com/hytseng0509/CrossDomainFewShot/blob/master/methods/LFTNet.py#L189
https://github.com/hytseng0509/CrossDomainFewShot/blob/master/methods/backbone.py#L207. This means when testing, the proposed method is the same as the base model. Is this right?

Owner

hytseng0509 commented Jul 21, 2020

Yes, the feature-wise transformation layers are used only during the training phase to improve the model generalization.

# for free to join this conversation on GitHub. Already have an account? # to comment